Simulating Quantum Many-Body Dynamics on a Current Digital Quantum Computer
Total Page：16
File Type：pdf, Size：1020Kb
www.nature.com/npjqi ARTICLE OPEN Simulating quantum many-body dynamics on a current digital quantum computer Adam Smith 1,2*,M.S.Kim1, Frank Pollmann2 and Johannes Knolle1 Universal quantum computers are potentially an ideal setting for simulating many-body quantum dynamics that is out of reach for classical digital computers. We use state-of-the-art IBM quantum computers to study paradigmatic examples of condensed matter physics—we simulate the effects of disorder and interactions on quantum particle transport, as well as correlation and entanglement spreading. Our benchmark results show that the quality of the current machines is below what is necessary for quantitatively accurate continuous-time dynamics of observables and reachable system sizes are small comparable to exact diagonalization. Despite this, we are successfully able to demonstrate clear qualitative behaviour associated with localization physics and many-body interaction effects. npj Quantum Information (2019) 5:106 ; https://doi.org/10.1038/s41534-019-0217-0 INTRODUCTION example, allowed us to study Hubbard model physics31,35 and 32–34 Quantum computers are general purpose devices that leverage many-body localized systems in two dimensions. More 1234567890():,; quantum mechanical behaviour to outperform their classical recently, there have also been advances in trapped ion quantum counterparts by reducing the computational time and/or the simulators, which have the benefit of being able to implement 6,7,37 required physical resources.1 Excitement about quantum compu- long range interactions, and have been used to study the tation was initially fuelled by the prime-factorization algorithm Schwinger-mechanism of pair production/annihilation in 1D developed by Shor,2 which is most popularly associated with the lattice QED7 and Floquet time crystals.38 ability to attack currently used cyber security protocols. More Universal quantum computers are also increasingly looking like importantly, it provided a paradigmatic example of dramatic a feasible setting for simulating quantum dynamics. One of the exponential improvement in computational speed when com- biggest advantages of using a quantum computer for this purpose pared with classical algorithms. It has since been realised that the is the flexibility it offers. A single quantum device could in potential power of quantum computers could have far reaching principle perform simulations that currently require several applications, from quantum chemistry and the associated benefits different experiments, using disparate methods. Furthermore, it for medicine and drug discovery,3 to quantum machine learning should be possible to access new physics not currently accessible, and artificial intelligence.4 Even before reaching these lofty goals, most notably, the simulation of lattice gauge theories with there may also be practical uses for noisy intermediate-scale dynamical gauge fields. These are ubiquitous in the theoretical quantum (NISQ) devices.5 study of strongly-correlated quantum matter but require multi- These quantum devices can be implemented in a large number body couplings, which have so far proven difficult to achieve in of ways, for example, using ultracold trapped ions6–11 cavity experiment.39,40 quantum electrodynamics (QED),12–15 photonic circuits,16–18 sili- A particularly exciting opportunity has been provided by IBM in con quantum dots,19–21 and theoretically even by braiding, as yet the form of an online quantum computing network called IBM Q. unobserved, exotic collective excitations called non-abelian any- This consists of a set of small quantum computers of 5 and 16 ons.22–24 One of the most promising approaches is using qubits that are available freely to the public, two 20 qubit superconducting circuits,25–27 where recent advances have machines accessible by IBM Q partners,41 and the qiskit python resulted in devices consisting of up to 72 qubits, pushing us ever API42 for programming the devices. The publicly available closer to realising so-called quantum supremacy.28 The apparent resources have already resulted in a spread of results, such as proximity of current devices to this milestone makes it timely to calculating the ground state of simple molecules,3,43 creating and review the current capabilities and limitations of quantum measuring highly entangled many qubit states,44,45 implementing – computers. quantum algorithms,46 48 and simulating non-equilibrium Richard Feynman’s original idea was to simulate quantum dynamics in the transverse-field Ising,49,50 Heisenberg51,52 and many-body dynamics—a notoriously hard problem for a classical Schwinger53 models, as just a few examples. Given the infancy of computer—by using another quantum system.29 Over the last quantum computing efforts, all these results are understandably couple of decades, this approach of using a purpose built small scale and of limited accuracy. quantum simulator has been extremely successful in accessing IBM is not alone in their efforts to make quantum computing physics beyond the reach of numerics on a classical computer. more mainstream, with Microsoft introducing the Q-sharp Most notable are cold atom experiments with optical lattices30–36 programming language,54 Google developing the Circq python where the natural evolution of the atoms corresponds to a high library,55 and Rigetti providing their own Quantum Cloud Service accuracy to that of a local Hamiltonian of choice. This has, for and Forest SDK built on Python.56 Rigetti and IonQ also provide 1Blackett Laboratory, Imperial College London, London SW7 2AZ, UK. 2Department of Physics, T42, Technische Universität München, James-Franck-Straße 1, D-85748 Garching, Germany. *email: [email protected] Published in partnership with The University of New South Wales A. Smith et al. 2 selective public access to hardware—based on superconducting qubits and trapped ions, respectively. All of these resources are allowing a lot of hands on experience with quantum computers from researchers and the general public all around the world. Furthermore, it highlights that there are currently several parallel efforts of research and development from industry focussed on quantum computing, quantum programming and cloud-based services that are flexible and prepared for future hardware. Given the current stage of development, and the immense expectation from the public and physics communities, it is timely to critically assess and benchmark the state-of-the-art. In this article we consider the far-from-equilibrium dynamics of global quantum quenches simulated on an IBM 20 qubit quantum computer. The models that we consider are of central importance to condensed matter physics and display a wide range of phenomenologies. By measuring a range of physical correlators, Fig. 1 Schematic of the implementation of Trotterized evolution. a The procedure of adding more Trotter steps to reach longer times. b, c we can assess the capabilities and limitations of current quantum are the basic and symmetric Trotter decomposition, respectively, for computers for simulating quantum dynamics. ^z ^ Àihj σ Δt ^ the Hamiltonian Eq. (3). In b the operators are Aj ¼ e j ; Bj ¼ σ^z σ^z σ^x σ^x σ^y σ^y Δ ^z Δt ÀiðU j j 1ÀJð j j 1þ ÞÞ t ^ Àihj σ ^ e þ þ j jþ1 and in c they are Aj ¼ e j 2 ; Bj ¼ σ^z σ^z σ^x σ^x σ^y σ^y Δt σ^z σ^z σ^x σ^x σ^y σ^y Δ ÀiðU j j 1ÀJð j j 1þ ÞÞ ^ ÀiðU j j 1ÀJð j j 1þ ÞÞ t RESULTS e þ þ j jþ1 2 ; Cj ¼ e þ þ j jþ1 . Setup In this article, we study global quantum quenches,57,58 that is we calculate local observables and correlators of the form These Hamiltonians are perfect testbeds for digital quantum simulation since they can be directly simulated on a quantum ψ ^ ψ ; ψ ^ ^ ψ ; (1) ðtÞ Oj ðtÞ ðtÞ OjOk ðtÞ computer – the spins-1/2 of the former correspond directly to the ^ where Oj are local operators and the time-dependent states are qubits of the latter, and the local connectivity of the qubits is 1234567890():,; suited for the local form of the Hamiltonian. ψ ÀiHt^ ψ ; jiðtÞ ¼ e jið0Þ (2) In our simulations we achieve the time evolution using a Trotter 66,67 and where jiψð0Þ is the initial state, which differs globally from an decomposition of the unitary time evolution operator ^ ÀiHt^ eigenstate of H^. In other words, we can consider jiψð0Þ to be the UðtÞ¼e . Our simulation proceeds by the following steps. ground state of a (time-independent) preparation Hamiltonian, First, we create the initial state, which is done by applying Pauli-X ^ ^ ^ ^ operations to the default initial state of the IBM machine, H0 ¼ H À hV, where V is a global perturbation, and the perturba- tion strength h is instantaneously quenched from zero at time jiÁ Á Á """ Á Á Á . Second, we discretize time and the evolution becomes a product of discrete evolution operations t ¼ 0. ^ ^ Δ ^ Δ We consider one dimensional spin-1/2 chains, consisting of N UðtÞ¼Uð tÞÁÁÁUð tÞ. Note that in our plots we include fi additional data points by varying δt in the final discrete evolution spins, initially prepared in either a domain wall con guration, Δ jiÁ Á Á ###""" Á Á Á , or Néel state, jiÁ Á Á "#"#" Á Á Á . Quenches from step. Third, we trotterize the discrete operators Uð tÞ into a these initial states are particularly easy to prepare due to the local product of one- and two-qubit unitaries. Finally, we decompose tensor-product structure of the initial state and have been studied these unitaries into the CNOT and single qubit rotation gates that in cold atom experiments.32 The time evolution after the quantum can be directly applied on the IBM devices. Please see Fig. 1 for a quench will be governed by a Hamiltonian of the form schematic of this procedure, and see Methods for more details.